Location: Crop Protection and Management ResearchTitle: A likelihood-based biostatistical model for analyzing consumer movement in simultaneous choice experiments Author
Submitted to: Environmental Entomology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/1/2014
Publication Date: 8/4/2014
Citation: Zeilinger, A.R., Olson, D.M., Andow, D.A. 2014. A likelihood-based biostatistical model for analyzing consumer movement in simultaneous choice experiments. Environmental Entomology. 43(4):977-988. doi: org/10.1603/EN13287. Interpretive Summary: Measures of consumer movement in food choice assays provide critical insight into consumer preference, and provide insights distinct from measures of consumption. Compared to measures of consumption, relatively little work has been done on the analysis of measures of movement from choice assays. Here, we describe a simple biostatistical model useful for analyzing repeated measures of consumer location in a paired-choice assay. With experimental data, the model can be used to estimate constant consumer attraction and leaving rates from two food choices; differences in choice-specific attraction and leaving rates can then be tested using model selection methods. We also describe methods to check that data meet critical assumptions of the model. Finally, using an example dataset with only a modest sample size, we illustrate that the model can be used to draw inferences on consumer preference and the mechanisms underlying preference.
Technical Abstract: Measures of animal movement versus consumption rates can provide valuable, ecologically relevant information on feeding preference, specifically estimates of attraction rate, leaving rate, tenure time, or measures of flight/walking path. Here, we develop a simple biostatistical model to analyze repeated measures of consumer location between paired food choices, under the special case of constant attraction and leaving rates. Our model focuses on the estimation of attraction rates and leaving rates for paired-choice assays. The model incorporates all available information on consumer movement to and from food choices during the trial, but does not require constant observation. We use experimental data on the movement of stink bug nymphs Euchistus servus, Nezara viridula between Heliothis virescens damaged and undamaged plants in a paired choice test to illustrate that the model can be used to draw inferences on consumer preference and the mechanisms (differences in attraction, tenure time or both) underlying preference. We found that attraction and leaving rates for N. viridula and attraction rates for E. servus did not differ between the two plant types. However, E. servus leaving rates differed between plant choices indicating that host acceptance behavior may be more important for determining preference than host selection behavior. The biostatistical model and related statistical analyses presented here should enable ecologists to obtain biologically relevant inferences from repeated measures of consumer movement in paired-choice assays.